Closed NicolasCazin42 closed 2 years ago
HI Nicolas,
Thank you for reporting the error. We reproduced it. However, this dataset version (with npz file instead of npy) was not supposed to be there! Because over the week end we uploaded the anonymized full size version of the dataset, we encourage you to download it again and this should be fully compatible with the current model version. Please, note that we moved from DropBox to Google, the download links have been updated too. Just in case, you can download the dataset using the following link: https://drive.google.com/file/d/1PpAcL5r2PRYMxDb46ASps9YqToL7uJcE/view?usp=sharing
Please, let us know if you experiment other issues and we would be happy to assist you Regards Julien
Hello,
I tried to test the provided code with the provided dataset, code and checkpoint as described in the README:
$ wget https://www.dropbox.com/s/y5kptmgxcngesp7/RADIal.zip
There is a compatibility issue between line 71 of file FFTRadNet/dataset/dataset.py) and the expected file extension (npy expected in the code but npz provided)
I found a quick fix by having a look at RADIal/loader/dataset.py) file, line 60. It consists in referring to npz file,
arr_0
file:With this fix, I am now able to launch 1-Train.py, 2-Test.py and 3-Evaluation.py scripts but more serious problems happened:
$ python 2-Test.py --config config/config_FFTRadNet_192_56.json --checkpoint FFTRadNet_RA_192_56_epoch78_loss_172.8239_AP_0.9813.pth
Launches an opencv window, camera images are updated, detected cars are always missed (bounding box always at the center of the image) whereas fft spectrum and freespace segmentation parts do not change over time. The the application crashed after several samples being displayed:$ python 3-Evaluation.py --config config/config_FFTRadNet_192_56.json --checkpoint FFTRadNet_RA_192_56_epoch78_loss_172.8239_AP_0.9813.pth
result in very small scores for a pretrained model:Do you have an explanation and a suggestion for using FFTRadNet properly ?
Best regards, Nicolas